Epithelial barrier dysfunction contributes to progression of intestinal and systemic disease. Long myosin light chain kinase isoform 1 (MLCK1) is trafficked to epithelial tight junctions in response to pathogenic stimuli, where it phosphorylates myosin II regulatory light chain (MLC) to reduce barrier function. Moreover, genetic or enzymatic MLCK inhibition prevents epithelial barrier loss in cell culture and animal models, suggesting its viability as a therapeutic target in inflammatory bowel disease (IBD), graft vs. host disease (GvHD), and other intestinal and systemic disorders. However, the myosin light chain kinase gene MYLK encodes both long and short MLCK, which share catalytic and calmodulin-binding regulatory domains, and therefore suitable reagents for MLCK1 inhibition in vivo are not available. We have identified several molecules, divertins, that can divert MLCK1 from the tight junction, limit stimulus-induced MLC phosphorylation, and prevent subsequent barrier loss in vivo and in vitro. Because divertins do not inhibit MLCK enzymatic activity, their toxicity profile should be excellent. Our strong preliminary data implicate FK506 binding protein 8 (FKBP8) as a specific binding partner of MLCK1 and that disruption of this interaction is the mechanism of divertin action. While identified divertins provide critical proof-of-principle, they require concentrations too high for therapeutic use. The goals of our study are i) to develop time-resolved, Frster Resonance Energy Transfer (TR-FRET) assays for screening protein-protein interaction (PPI) inhibitors in vitro and inside living cells; and ii) to use these assays o identify novel inhibitors of MLCK1/FKBP8 interaction that will both advance biological research and serve as preclinical trial lead agents. The objective is to demonstrate high signal-to-background ratio (SBR) and Z'- factor of the proposed assays and to apply these assays to screen an exceptionally diverse chemical library for MLCK1/ FKBP8 inhibitors. Our central hypothesis is that the well established sensitivity (SBR > 50) and reproducibility (Z' > 0.85) of cell-free TR-FRET assays can be applied to screens within living cells. This hypothesis rests on strong preliminary data showing that i) luminescent trimethoprim-terbium complex conjugates specifically label E. coli dihydrofolate reductase (eDHFR) fusion proteins in vitro and in live cels and interactions between eDHFR and green fluorescent protein (GFP) fusions can be sensitively detected as terbium-to-GFP TR-FRET. The hypothesis will be tested via specific aims: 1) To develop a cell-based TR- FRET HTS assay using a model interaction - FKBP12/FRB; 2) To identify inhibitors of MLCK1/FKBP8 in cell- free assays and intact epithelial monolayers; and 3) To counter-screen identified lead compounds in vitro, in live cells, and in live mice. The rationale for this innovative approach to non-enzymatic kinase inhibition is that it holds great promise for barrier restoration, and even greater potential as a new paradigm for targeting kinases and enzymes with improved specificity. In addition, the TR-FRET approach will yield robust high throughout screens to identify agents that modulate a wide variety of PPIs in live cells.